Re: [Manycouches] Webex configuration

[Toerless Eckert <[e-mail]>]

Cc'ing manycouches

On Fri, Jun 12, 2020 at 08:45:10PM +0200, Carsten Bormann wrote:
> Good point that needs to be added to the Webex FAQ (once we write one up).
> 
> It is amazing to see the difference between Webex and a system that was designed by people who actually do have meetings not unlike the ones we have: BigBlueButton (BBB) that is.  It has polls (really useful in a classroom setting), you can upload your slides (so they don???t need to be sent as video) and draw on them in realtime (and, optionally, have participants draw on them as well); when integrated with an LMS (learning management system) you even have automatic identification.

Let me Cc fluffy explicitly, he should know more authoritatively,
 i actually would like to phrase it more as questions:

Webex has polls, can upload slides, allows drawing on realtime... not sure about the other features.
BUT: these features don't seem to work in the RTCweb == browser version.
Michael Richardson i think would step up here and ask not to rely on features
not supported in the RTCweb version (whatever the tool is). And i am on the
fence with that argument.

Carsten: Do you know if these features work in BBB in RTCweb or equally only in native clients ?

I remember Cullen had a bix of an exchange about feature limits in RTCwebs
earlier on the IETF mailing list, so i am wondering and curious whether that
argument applies to these features. I could only imagine that the slides and
drawing could relate to RTCweb limitations. The raising hands sounds like
solely part of the javascript app.

If there are reasonable problems to do advanced features via RTCweb,
whatever the conferencing product is we choose, then i would suggest to reconsider
the "Must rely on lowest-common-RTCweb-denominator".

> Re the mike challenge: Everybody joining a BBB conference who intends to be able to speak (not a given with BBB - there is a listen-only mode) needs to go through a quick mike/speaker test.  Each time.  And it really only takes two seconds.

I don't think these tests find really common problems. Yesterday on a university
zoom meeting one of the speakers was continuously broken up, sounded like
very reliably overloaded upstream link from home. The tests typically are
only level issue based.

> The potential to have your system designed by someone who knows what is important in real life (and who actually cares) is one of the really big benefits of the empowerment WebRTC gave us.

See above Q.

Cheers
    Toerless

> But yes, if you need to use a system that doesn???t have this, do chat in the [-4 min, +4 min] interval.  When we did the OneDM spring conference (with several two-to-three hour segments per week), we actually started to make a point out of using that time for informal chatting, as well.  > Of course, it helps if the recording hasn???t been started yet :-)
> 
> Grüße, Carsten

-- 
---
[e-mail]